Networks and the Myth that Flatter Organizations are Better

by cv harquail on January 15, 2010

Are flatter organizations really “better”? If they are better, how?

Hey, I already wrote a dissertation, so I’m not going to take on that question in its entirety. And, I’m not going to do the proper academic thing of being super-specific and qualifying my points. You got complaints? Email me and I’ll send you the scientific citations. Or, I’ll chair your dissertation. < grin >

Otherwise, bear with me here. I want to re-consider a really important assumption about one way that flatter organizations with internal network structures are better. (I’m thinking about the organizations advocated by folks bringing social media inside organizations, mostly proponents of Enterprise 2.0 and social business.)

Flatter, more networked organizational structures do not significantly reduce power inequalities among employees or across domains within a firm.

Just last week I was complaining that proponents of socially-mediated organizations aren’t being radical enough. After all, these new work arrangements and organizational structures can really change the world!

201001141711.jpgThink about it—the ideal network structure and work processes of Enterprise 2.0 look an awful lot like feminist organizations. And, we are already seeing emergent networks of social advocates that demonstrate more collaborative, more egalitarian dynamics. So what’s the problem?

The problem is that these kinds of structures, when brought into your basic business organization, don’t necessarily bring along with them a real change in employee relationships within the firm. The change to a more networked structure can make an organization more productive, but it doesn’t make the organization more egalitarian, more democratic, or more just.

A bummer, I know.

Studies show that organizations that are flatter because they have a network structure encapsulated or embedded inside them still, in the big picture, feel and act like hierarchies (Dean, 2007).

Oh, sure, you as an employee feel a bit freer in your day-to-day situation (especially if you have some control over your time). But, overall, you and your colleagues are still locked into a bureaucratic-ish organization where authority over medium and large-scale decision remains concentrated in a few high levels of (management) employees.

Flatter does not mean that power is more evenly-distributed across the levels that remain.

In a “flatter” organization, there are fewer levels of decision-making authority (e.g., less hierarchy) than there were before. We assume that when levels are reduced, some amount of decision-making power is freed up. (For example, if we get rid of the brand supervision level, someone else at some other level gets to choose the new label while someone else sets cost targets.)

Further, we assume that if a level is removed, the decision-making power of that level gets evenly distributed across the remaining levels. For example, if the organization drops from 5 levels to 4, the power once held by the eliminated level gets equally distributed across the remaining 4 levels. Everybody gets 25% more decision-making power. More power to us, less power over us.

Our assumptions are wrong. It doesn’t work that way. Power is rarely redistributed in any kind of egalitarian fashon. A little power might go to the levels below the ones eliminated, but the important power stays up above.

Although power gets redistributed in a network, the surrounding hierarchy doesn’t actually give up power that matters.

When organizations restructure some units into to networks, they are usually very strategic about what ‘power’ and ‘authority’ is delegated to the network or team.

Networks/teams get more “production-level authority” over who’s doing what within the overall project, what parts of the day are spent where, and the like. But the team or network doesn’t get ‘high level’ decision making authority. This still remains with upper management.

Even when managers in the hierarchy above the network solicit input and invite innovative ideas, ultimately it is the managers (still) in the hierarchy that make the big decisions. Authority is still concentrated in higher-level managers, who make the important decisions, decisions about whether there will be layoffs, how much money goes into everyone’s 401Ks, whether the project is outsourced, etc.

Flatter may mean more power over your immediate situation, but still the same (low) amount of power over the big picture, adding to a minor net reduction in power difference.

A bummer, I know.

201001141713.jpgSome people are going to argue that networked organizations really do have different internal power dynamics that do traditional hierarchies. That’s true, and sometimes the degree of shared power is really significant. In fact, given all this additional autonomy and collaboration and input-giving, employees might not even notice that they still lack power where it matters the most: over the distribution of gains.

Keep in mind the ‘real’ business reason that organizations restructure and create internal networks. Organizations restructure to improve productivity. They want more stuff produced and they want to produce it at a higher quality. Why? So that the organization is more profitable.

The gains of a flatter, more networked structure are rarely distributed in an egalitarian way.

Consider where those productivity gains go– into “surplus value”, otherwise known as profit.

When the employees of an organization become more productive because they feel more autonomy over their work, because they have more input into decision-making, and because they are able to collaborate with less friction, where do these profits go? Are they evenly distributed across the layers of employees whose work created this extra value? Because if gains were distributed this way, if would demonstrate quite clearly that the organization was more egalitarian in a material way (pun intended).

I’m not saying that, in order to be more egalitarian, everybody in the organization has to get paid the same. I’m arguing that in an organization that is “flatter”, where there is more democracy, more autonomy and more decision-making power for employees, we would see all employees benefiting financially at least at similar rates.

Go ahead, call me Dr. Buzzkill. Bringing some management science into the picture does refocus things, doesn’t it.

The overall point is, creating networks inside organizations won’t necessarily make these organizations more egalitarian, more democratic, and better for everyone.

201001141709.jpg We can use network structures, shared decision processes, and collaborative work systems to make organizations more just, if we do this intentionally. On purpose. With purpose.

We just have to make egalitarianism and justice overarching goals. They have to be as important, if not more important, than increased innovation, nicer interpersonal interactions, and yes more surplus value.

Caveats include:

Organizations starting from scratch (i.e., greenfields) find it easier to create egalitarian structures, though research shows that these structures can find it harder to sustain legitimacy, depending on their institutional environment.

Organizations that are organized around a mission or purpose (e.g., non-profits, ideological organizations) often have core values that override hierarchical power hoarding.

Yes, I am implicitly explicitly saying that organizations that are more democratic, more egalitarian, and more just are ‘better’. No, I am not saying that all organizations should be flat, or that hierarchy should be abolished.

More to come.

Joan Acker, 2006. Inequality Regimes: Gender, Class, and Race in Organizations, Gender & Society 2006; 20; 441- 464.

Conaldi, Guido. 2009. Flat for the few, steep for the many: Structural cohesion as a measure of hierarchy in FLOSS communities. Working paper. Institute of Management, University of Lugano, CH-6904 Lugano, CH.

Dean, Paul. 2007. Flat and Egalitarian? Evaluating worker hierarchies in software companies. Unpublished Master’s Thesis.  University of Maryland, College Park, MD.

Rajan, Raghuram G. & Wulf, Julie, 2006. “Are perks purely managerial excess? Journal of Financial Economics Elsevier, vol. 79(1), pages 1-33, January.

Photos from Flickr:
Ladders and Lamp
from tomswift46
Jacob’s Ladder  from
Jacob’s Ladder from ShellyS


Jon Ingham January 17, 2010 at 10:01 am

CV, I think this is absolutely right. Flatness and networking won’t make any real difference at all (ie to the feel of an organisation vs its productivity etc) unless there’s a strong intent to do so (ie for it to be different). The intent – the overarching goals – is the important thing.
.-= Jon Ingham´s last blog ..The Social Revolution – isn’t hierarchy to networks =-.

cv January 19, 2010 at 10:34 am

Jon, I think you’re right to describe the challenge as getting up the ‘mojo’ to assert that simple systems changes (1) aren’t simple, and (2) need to be intentional around issues like power distribution, decision-making, and relationship values.
I’m realizing that we both think the other is right. So it’s time to collude. I’ll muster up my mojo if you’ll rally yours to the cause.

Jordi February 20, 2010 at 6:51 pm

First, CV, you are right to point out that a change in structure by itself will not lead to more participation or more shared prosperity. I think the hopeful rhetoric that suggests such out comes conflates structural change with holistic change.

Second, a flattened organization that devolves decision-making without appropriate information sharing and mechanisms for interacting can produce a lot MORE conflict. Right? If lots of teams and groups are jockeying for key information or organizational resources, or merely muddling through unstructured problems, then we might see more conflict. That is not necessarily a bad thing, but some may imagine that network organization rhetoric implies a seamless kind of collective mind and coordination.

Third, lots of network theorists, like me, :<), and Castells, are quite explicit that viewing the world as networks does not sublimate or dissolve power. The opposite: networks have their own "programs" that dictate inclusion or exclusion. Whoever writes or inserts the code has power. In the case of your networked organization, it may be the managers (or the nebulous "market") who have the power to determine what the network values and hence who is in or out of the network.

cv harquail February 23, 2010 at 10:21 am

Hi Jordi-

Thanks so much for your comment!

You are so right– the rhetoric involved in conversations about networks obscures the real challenges that networked structures pose. It’s an interesting case of conceptual inversion– if power = bureaucracy = hierarchy, and hierarchy is the opposite of networks, then networks have no bureaucracy and no power issues. But people forget that the transitive property works better in algebra than it does in human social systems.

Do you know of any work (or have you done some yourself?) that sets out the specific power-related challenges of a network structure, in a really simple way?

Comments on this entry are closed.